Semiconductor pixel unit for sensing near-infrared light, optionally simultaneously with visible light, and a semiconductor sensor comprising same

Abstract

A semiconductor pixel unit for sensing near-infrared light, and for optionally simultaneously sensing visible light. The pixel unit comprises a single substrate with a first semiconductor region and a second semiconductor region electrically separated by an insulating region, for example a buried oxide layer. The pixel unit is adapted for generating a lateral electrical field in the second region for facilitating transport of photoelectrons generated in the second region by near-infrared light passing through the first region and the insulating region.

Description

FIELD OF THE INVENTION[0001]The invention relates to the field of semiconductor structures for near-infrared (NIR) light and optionally simultaneously sensing visible light. More in particular, the present invention relates to an imaging structure (e.g. a pixel unit) capable of simultaneously capturing visible light information (e.g. grayscale information or color information) and near-infrared light information (e.g. representative of distance information also known as “time-of-flight” (TOF) information).BACKGROUND OF THE INVENTION[0002]Solid state silicon image sensors became ubiquitous in the recent years. The first high quality image sensors were fabricated in a CCD technology which is nowadays being more and more replaced by CMOS technology. While typical CMOS based image sensor sensitivity is still lower than CCD based sensors, the integration advantages that CMOS technology can provide make it a technology of choice for today's image sensors. Besides the integration, CMOS tec...

AI Technical Summary

Benefits of technology

The present invention aims to improve the detection of near-infrared light in semiconductor devices and pixel units. It achieves this by using a filter that enhances the demodulation contrast of the light, making it easier to distinguish the desired signals. The use of this filter has advantages in applications such as time-of-flight, where it helps to remove most of the unwanted signals and improve the accuracy of the resulting information.

Problems solved by technology

For machine vision applications, despite increasing resolution, increasing dynamic range of image sensors and increasing computational power of CPUs, it is impossible for a machine to extract distance information from a single 2D-image.

However, this requires two cameras and a powerful processor.

Although it is possible to get an image from a (pure) time-of-flight camera, the quality of such an image would be much inferior to the image quality of a typical image sensor as used e.g. in digital cameras.

Another problem is that image sensors and time-of-flight cameras set different requirements towards the optics.

However this requires a high system cost.

However, the cost of such a sensor is relatively high.

Images